Gradient descent with momentum based neural network pattern classification for the prediction of soil moisture content in precision agriculture

Irrigation in agricultural lands plays a vivacious role in water and soil conservation. Future prediction of soil moisture content using real-time soil and environmental parameters may provide an efficient platform for agriculture land irrigation requirements. In this paper we have proposed one optimization technique like Gradient Descent with Momentum is used to train neural network pattern classification algorithm. The algorithm is tested for the prediction of soil moisture content in each one hour advance by considering eleven different soil and environmental parameters collected during a field test. The prediction errors are analysed using MSE (Mean Square Error), RMSE (Root Mean Square Error), and R-squared error. The performance of the proposed soil moisture content prediction is well documented in this article.